The goal of this K01 proposal is to provide John Monterosso, Ph.D., with the training and support to become an independent investigator in the field of clinical substance abuse research. The research plan builds on the candidate's expertise in animal models of impulsivity; its basic aim is to measure analogous constructs in cocaine dependent patients and to use these measurements in conjunction with cue reactivity to better characterize patients and to better predict treatment outcomes. The training plan provides extensive clinical exposure, individual mentoring in areas related to the research, and structured didactics. Both the research and training plans are opportunistic, utilizing readily-available patient populations from ongoing research, and seasoned investigators in the University of Pennsylvania research community. Dr. Anna Rose Childress will provide primary mentorship and daily supervision in the proposed research; her expertise lies in the phenomenology of cocaine dependence, cue-induced craving, and its brain substrates. Dr. Ronald Ehrman will provide expertise in the laboratory measurement of cue reactivity, and in statistical modeling. Dr. Charles O'Brien will provide expertise on the neurobiology of addiction, will ensure support for successful execution of the project, and will chair an Advisory Board of selected investigators who will be a resource to the candidate and monitor his development. Relapse is a cardinal feature of the addictions, and the one which exacts the greatest human and economic costs. Understanding its mechanisms is critical to reducing these costs. Though cue-induced craving and arousal have been offered by us, and by others, as one possible mechanism, craving has been an imperfect predictor of drug use: not every craving episode eventuates in relapse, and patients vary in their ability to manage these episodes. Host variables may help explain this variability. Since the reward of drug is immediate and the benefits of abstinence are delayed, individual differences in sensitivity to future consequences may be an important variable. Tasks have recently been developed for assessing this dimension: one derived from an animal model of impulsivity, and two others from neuropsychological research with orbitofrontal patients, who show extreme behavioral "myopia" (Damasio, 1994). We have combined these methods into a "Myopia Battery (MB)" for use in the proposed studies. We have conducted a large pilot study with encouraging results. Differences in myopia may be particularly useful for understanding the disconnect between reported craving and drug use/relapse. In Study 1, cocaine-dependent patients participating in a large-scale treatment outcome study (n= 120) will be assessed on cue reactivity/craving, ASP, Impulsivity, I.Q. and will be administered the MB. The MB will also be administered to a group of matched controls (n=80). In Study 2, the MB will be administered to cocaine patients (n=60) participating in ongoing PET studies that directly assess orbitofrontal activity. Our primary hypotheses are: 1) cocaine-dependent patients will perform worse on the MB than controls; 2) performance on the MB will be especially poor in cocaine-dependent patients with ASP; 3) performance on the MB will be related to the variability of patients' ability to manage craving states without relapsing, and 4) MB performance will be correlated with orbitofrontal functioning. The link between severe myopia and orbitofrontal impairment is particularly exciting given emerging evidence (including that collected recently in our own lab) of orbitofrontal deficits in cocaine addicts relative to controls. Whether the dysfunction is a predisposing factor for, or a consequence of, stimulant use - or both - it could undermine an important psychological resource for recovery.